Foreign material contamination detection
View Patent ↗There are provided a system, method and computer program product for detecting foreign materials in a semiconductor manufacturing process. The manufacturing process uses a plurality of semiconductor manufacturing tools. The system categorizes at least one monitoring wafer according to one or more categories. The system supplies the categorized monitoring wafer to a semiconductor manufacturing tool. The system observes a level of contamination on the categorized monitoring wafer. The system compares the level of contamination to a threshold. The system cleans the tool in a response to determining that the level of contamination is larger than the threshold. The system determines which category of the wafer leaves a highest level of contamination on the tool. The system identifies a root cause of the highest level of contamination on the tool.
1. A method for detecting foreign particles in a semiconductor manufacturing process, the manufacturing process using a plurality of semiconductor manufacturing tools, the method comprising:
categorizing at least one monitoring wafer according to one or more categories;
supplying the categorized monitoring wafer to a semiconductor manufacturing tool;
observing a level of contamination on the categorized monitoring wafer;
comparing the level of contamination to a threshold;
cleaning the tool in a response to determining that the level of contamination is larger than the threshold;
determining which category of the categorized monitoring wafer leaves a highest level of contamination on the tool; and
identifying, based on the determined category, a root cause of the highest level of contamination on the tool,
wherein a processor performs one or more of: the categorizing, the supplying, the observing, the comparing, the cleaning, the determining, and the identifying.
2. The method according to claim 1 , wherein the level of contamination on the monitoring wafer represents a number of foreign particles on the tool.
3. The method according to claim 1 , wherein a category of a monitoring wafer is determined by semiconductor manufacturing tools used.
4. The method according to claim 1 , wherein the determining the highest contamination category includes solving best fit problems.
5. The method according to claim 1 , wherein the identifying includes:
searching for a category with the highest level of contamination in a dramatic circumstance;
observing a change in a category with an average level of contamination in a usual circumstance; and
distributing new monitoring wafers in the category with the highest level of contamination or in the category with the average level of contamination or in both categories.
6. The method according to claim 1 , further comprising:
modeling the level of contamination on the tool.
7. The method according to claim 2 , wherein the foreign particles are observed by a microscope.
8. The method according to claim 4 , wherein the best fit problems include one or more of: Least Square method and Maximum Likelihood method.
9. The method according to claim 4 , wherein a solution of the best fit problems classifies parameters of the level of contamination of the tool.
10. The method according to claim 6 , further comprising:
using at least one statistical distribution to create the model.
11. The method according to claim 9 , further comprising:
dividing the parameters into two groups according to a variable;
calculating a first coefficient based on the variable and a number of the parameters;
comparing the first coefficient with a reference coefficient;
evaluating whether the variable generates a largest deviation between the first coefficient and the reference coefficient; and
deciding that the two groups is a best classification of the parameters in a response to determining that the variable generates the largest deviation.
12. The method according to claim 10 , wherein the statistical distribution includes one or more of: Poisson distribution and Bernoulli distribution.
13. A system for detecting foreign particles in a semiconductor manufacturing process, the manufacturing process using a plurality of semiconductor manufacturing tools, the system comprising:
a memory device;
a processor connected to the memory device;
the processor performs steps of:
categorizing at least one monitoring wafer according to one or more categories;
supplying the categorized monitoring wafer to a semiconductor manufacturing tool;
observing a level of contamination on the categorized monitoring wafer;
comparing the level of contamination to a threshold;
cleaning the tool in a response to determining that the level of contamination is larger than the threshold;
determining which category of the categorized monitoring wafer leaves a highest level of contamination on the tool; and
identifying, based on the determined category, a root cause of the highest level of contamination on the tool.
14. The system according to claim 13 , wherein the level of contamination on the monitoring wafer represents a number of foreign particles on the tool.
15. The system according to claim 13 , wherein a category of a monitoring wafer is determined by semiconductor manufacturing tools used.
16. The system according to claim 13 , wherein the determining the highest contamination category includes solving best fit problems.
17. The system according to claim 13 , wherein the identifying includes:
searching for a category with the highest level of contamination in a dramatic circumstance;
observing a change in a category with an average level of contamination in a usual circumstance; and
distributing new monitoring wafers in the category with the highest level of contamination or in the category with the average level of contamination or in both categories.
18. The system according to claim 13 , wherein the processor further performs step of:
modeling the level of contamination on the tool.
19. The system according to claim 14 , wherein the foreign particles are observed by a microscope.
20. The system according to claim 16 , wherein the best fit problems includes one or more of: Least Square method and Maximum Likelihood method.
21. The system according to claim 16 , wherein a solution of the best fit problems classifies parameters of the level of contamination of the tool.
22. The system according to claim 18 , wherein the processor further performs step of:
using at least one statistical distribution to create the model.
23. The system according to claim 21 , the processor further performs steps of:
dividing the parameters into two groups according to a variable;
calculating a first coefficient based on the variable and a number of the parameters;
comparing the first coefficient with a reference coefficient;
evaluating whether the variable generates a largest deviation between the first coefficient and the reference coefficient; and
deciding that the two groups is a best classification of the parameters in a response to determining that the variable generates the largest deviation.
24. The system according to claim 22 , wherein the statistical distribution includes one or more of: Poisson distribution and Bernoulli distribution.
25. A computer program product for detecting foreign materials in a semiconductor manufacturing process, the manufacturing process including a plurality of semiconductor manufacturing tools, the computer program product comprising a storage medium readable by a processing circuit and storing instructions run by the processing circuit for performing a method, the method comprising:
categorizing at least one monitoring wafer;
supplying the categorized monitoring wafer to a semiconductor manufacturing tool;
observing a level of contamination on the monitoring wafer;
comparing the level of contamination to a threshold;
cleaning the tool in a response to determining that the level of contamination is larger than the threshold;
determining which category of the categorized monitoring wafer leave a highest level of contamination on the tool; and
identifying, based on the determined category, a root cause of the highest level of contamination on the tool.